Jump to content

"strikeouts Don't Matter"


K26dp
 Share

Recommended Posts

I completely agree on the offense portion. I used to scoff when metrics folks tried to tell me that there were no indications that strikeouts impact overall offensive performance. That fails the laugh test. Any amount of common sense would display that simply by putting the ball in play, the odds of good things happening increase dramatically.

With regards to the team, I posted the list of who we lost and who we gained several pages ago. It's absolutely brutal on paper. The only reason I am not freaked out about it is that I am confident Hart has plans to fill the holes he created in dealing MLB players for prospects.

This is a promised response to this topic from the Braves Off-Season Thread. This is long, and I didn't want to clog up the normal talk with this.

TL;DR version: strikeouts do matter, but probably not in the way most people think. The average value of a productive out isn't that high. Hitters can still be productive, even if they strike out a good bit. Hitters that don't strike out a lot can be unproductive.

Long version:

“Strikeouts don’t matter”.

These three words to come out of the sabermetric analytic revolution are probably the most misunderstood when it comes to evaluating the performance of hitters. It’s become a certainty with some of the analytically-inclined, taken at face value. To the rest of the baseball world though, it’s preposterous. Of course strikeouts matter. If you’re a Braves fan, all you really need to do it point to B.J. Upton and say “if strikeouts don’t matter than this guy is an All-Star.”

As so often the case, this is a nuanced idea that over time has gotten whittled down to shorthand. Proponents don’t explain it very well. Opponents use it as a punch line for everything wrong with advanced analytics.

So what is the deal?

First, I want to set the groundwork by stating some important facts. These are not controversial, and should be understood by anyone who watches the game.

· The purpose of the offense is to score runs.

· The offense can theoretically score any amount of runs before 27 batters are retired.

· Therefore the “clock” for the offense is 27 outs.

· Therefore the most basic positive outcome of a batter’s plate appearance is to not get out; by avoiding getting out, the offense gets more opportunities to score runs.

· Any outcome of a plate appearance where the batter did not get out is positive for the offense, but some outcomes are better than others. In all cases, the most positive outcome is to hit a home run, as it scores not only the batter, but all other baserunners.

· Any outcome of a plate appearance where the batter is retired is negative for the offense, but some outs are not as bad as others. Outs that advance a base runner are called “productive outs”. The best of these, a sacrifice that scores a runner, can actually be a positive for the team under certain circumstances.

· Hitting professional pitching is really, really hard, even for professional hitters. Even the best fail @60-65% of the time. Most fail @68-70% of the time. If a hitter consistently fails at a greater clip than that, they likely will not continue to be employed as a professional hitter, unless some other part of their game is extraordinarily good.

LINEAR WEIGHTS

A strikeout (K) is one of five basic ways the defense can get the batter out. The other four are groundball (GB) outs, infield fly (IFFO) outs, outfield fly outs (FO), and lineouts (LO). Many pitchers strive for strikeouts because there is almost no chance for a defensive miscue causing a batter to be safe.

When the bases are empty, as far as the offense goes all outs are essentially the same. In these cases, strikeouts don’t matter, because they are no better or worse than the other types of outs; the result of the plate appearance is failure.

When there is a runner on base, however, some outs can be better than others. The question then becomes, how much valuable is one out over another in those situations?

In 1984, baseball writers and early sabermatricians Pete Palmer and John Thorn wrote the book The Hidden Game of Baseball. In this book, Palmer outlines an evaluation system he called “Linear Weights”. The theory behind the Linear Weights System is to objectively give a value to each individual outcome of a baseball play. Palmer determined after exhaustive analysis that each event can be given a value that positively or negatively impacts a team’s chance of winning or losing a particular baseball game, and that these values remain extremely consistent over the course of baseball history. Each event is weighted against the likelihood of a team scoring a run after the event occurs; this is called Run Expectancy. A cousin to this is Win Expectancy.

The idea that a baseball game can be broken down into discrete results and each result can be given a value that is a ratio to the likelihood that the event will contribute (or inhibit) a team win is essentially the basis of all sabermetric thought. WAR is essentially this concept brought into one number and assigned to a player.

So a strikeout, like other outs, lowers Run and Win Expectancy roughly the same. However, other outs, depending on the game state (base-runner, number of outs, score) can alter Run and Win Expectancy differently.

PRODUCTIVE OUTS: AN OXYMORON

Productive outs occur when the game state changes to favor the offense after a batter is retired. This usually involves advancing a baserunner.

The problem with “productive outs” is that they are often not really that productive. Advancing a baserunner is nice, but not getting out is the better result. Managers that purposely give up outs in order to advance baserunners (like using sacrifice bunts) are often derided by analytically-included fans and pundits, because the Run and Win Expectancy almost always goes down after such an out. Even the best productive out, the sacrifice fly that scores a run from third base, has its usefulness modified by the game state; for example, it’s not much good in the 9th inning, the batting team down by three runs, and one out. After the sac fly, the offense is down by two runs, but they have two outs and the bases empty. Even though the play was a net positive Run Expectancy play, the Win Expectancy actually went down. The value of the run is less than the value of the out.

On average, the Run Expectancy of a productive out is approximately 0.167 runs, which is a very, very small amount. To put it in perspective, a single is worth 0.47, a walk or HBP is 0.33, and a home run is 1.4.

So a productive out is obviously better than an unproductive one, but to such a small degree that it rarely out-values the negative impact of the out itself.

Of course, what frustrates fans (and some inside baseball people as well) is that when a batter strikes out, he doesn’t give himself even the opportunity to have a productive out, or even perhaps reach on an error. And there’s the reality that a strikeout completely negates the potential to get a hit.

This is where it’s important to understand that simply striking the ball with the bat doesn’t guarantee anything at all, how well the ball is struck is important.

“PUT THE BALL IN PLAY AND GOOD THINGS HAPPEN”

This is a phrase used ad nauseum by Braves broadcasters Chip Carey and Joe Simpson. Of course, when know this is not always true; a strikeout for example is a preferable outcome to hitting into a double play. Of course, you can’t have a hitter not ever swing the bat when there’s a runner on first base for fear of a double play. Or can you? It depends on the hitter of course.

The easiest example is this is when the pitcher hits with a man on first base. A pitcher hitting is an extreme example that essentially breaks the Linear Weights System; the system assumes a professional hitter, and most pitchers are not. Bunting the runner to second base is a perfectly reasonable outcome of a pitcher plate appearance.

But what about real hitters? Here is a chart that shows hitter production on each type of batted ball:

BATTED BALL TYPE  AVG       ISO     wOBAGround Ball      .239      .020    .220Fly Ball         .207      .378    .335Line Drive       .685      .190    .684

(Source FanGraphs)

This chart shows in numbers what most fans understand intuitively; the harder you hit the ball, the higher the likelihood that there will be a good outcome. Good things do happen to hitters that put the ball in play… if they hit it hard. Most of the time when that happens, they’ll get a hit. (Aside: Infield Fly Balls aren’t on the chart because they are essentially as useless as strikeouts; they are converted to outs at a near 100% clip. Even in the case of an error with less than two outs, the Infield Fly Rule will make it so that baserunners can’t advance.)

However, bad things can happen to hitters that put the ball in play on the ground with a runner on first. It’s more likely such a ball will be converted into two outs. This is where the hitter needs to understand the game state, and adjust his approach accordingly. It’s not enough just to put the ball in play, he must hit the ball with authority. The batter will want to make sure he only swings a pitches he can handle, and take a base-on-balls if the pitcher won’t throw strikes. Pitch recognition is very important. And if the pitcher makes good pitches, it’s better to strikeout than swing weakly. Yes, there’s a chance the defense doesn’t make a play on that ground ball, but it’s unlikely (but when that does happen, you can count on Chip to make sure the audience knows that “good things happen when you put the ball in play”).

“STRIKEOUTS DON’T MATTER”

So not all strikeouts are really made equal.

Some hitters strike out because they don’t see the ball well, or get fooled. They swing at pitches outside the strikezone and look at pitches inside the strike zone.

Some hitters strike out because they are taking a lot of pitches, looking for a ball to drive, and making the pitcher work for the out.

In the end, the discrete results may be the same, but which hitter is more likely to have a positive outcome his next plate appearance?

When someone says “strikeouts don’t matter”, they are really talking about the second type of batter. These kinds of batters may have high strikeout rates, but provide value because they get on base a lot (that is, they are good at not getting out in general), or hit for power, or run the bases well, or provide great defense. These types of players are All-Stars and team MVPs, even with high strikeout rates.

The first type of batter is a problematic, because his batting approach is not allowing him to be successful. Chances are, that hitter is not doing other things well either, such as getting on base by walking, since they aren’t going deep in the count. Even if they are good baserunners, that value is diminished by not getting on base. Hopefully they are playing outstanding defense or have an immovable contract, because otherwise they probably won’t be employed for long. These types of players are… well, it’s B.J. Upton.

In these cases strikeouts “matter”, but mostly in the sense of they are getting out too much in general. In fact, there is likely a strikeout rate ceiling where even with positive contributions elsewhere are brought down too much by the strikeout rate. Research is ongoing to determine that ceiling, but it’s probably around 25%.

On the flipside, it doesn’t matter if you don’t strikeout a lot, if you aren’t doing anything otherwise to help the team offensively. Braves shortstop Andrelton Simmons is an example of this. Despite ranking 6th in MLB in putting the ball in play last season, he only had a .244 batting average. He only walked at a 5.6% rate as well, so his OBP was a pretty terrible .286. You probably won’t be surprised to see his batted ball breakdown either:

BATTLED BALL TYPE       PERCENTAGE OF BALL IN PLAYGround Ball                       52.4Fly Ball                          31.2Infield Fly Ball                  10.8Line Drive                        16.4

In essence, it doesn’t matter how much a hitter puts the ball in play if they hit so few line drives. For Simmons to improve his production he needs to hit the ball with more authority. He will likely not achieve this until he becomes more selective as a hitter. Doing this will also have the side effects of increasing his walks and increasing his strikeouts.

This is also way sabermetric types can say “strikeouts don’t matter”: in the cases of many batters, increased strikeouts are a sign of a more selective hitter, and can accompany an increase in production.

And thus we come to the heart of the paradox. Sometimes a hitter will become more productive when they strikeout more.

STRIKEOUTS DO MATTER

Unfortunately, even those that understand the paradox often have a hard time explaining it (and I am under no delusions that I’ve done any better). At the same time, fans (and broadcasters) that already look sideways at “number crunchers wasting time in their mother’s basements yada yada yada” latch on to “strikeouts don’t matter” as proof positive that those folks should try watch more games than sitting in front of a computer, or whatever.

These folks also tend to overvalue the “productive out”, which usually isn’t that productive… but sometimes will be the difference between winning and losing a game. Its human nature to remember that time that your team won with scrappy small-ball, and forget all the times that that runner that was moved over with an out and was stranded. It’s easier to recall that time a sacrifice fly won the game, rather than the times it was just the second run in a game that was decided by five.

Nevertheless, I believe strikeouts do matter, because I think you learn as much about a hitter by how he strikes out as by how he gets on base. The Braves have perfect, extreme illustrations of this in the Upton Brothers. Both strike out at a rate greater than average, but how they strike out cannot be more different.

Edited by K26dp
Link to comment
Share on other sites

BRAVO!!!!! Thank you for the great post K26dp. I know there are more than a handful of posters here that won't bother to read and understand what you just wrote but I appreciate the great time you took to put this together for those of us that do understand this.

Also I want to go back to that last part when you mentioned B.J. Upton. An example of the other type of hitter you were talking about is Jason Heyward. That's where the disconnect occurs for the posters here like you, me, and ATLBrave that truly understand the offensive value of Jason Heyward and the posters here that think Heyward is a bad or average offensive player.

Link to comment
Share on other sites

As I said in the other thread, the problem with this mentality is that when a front office goes all in with it it will lead to something like what the Braves have had for the past two seasons with a feast or famine lineup and players going into sudden decline. The high K rate is overlooked by most when they're hitting for power (2013) because they put up flashy numbers and win at a higher rate but when they do not offensive production collapses and the losses pile on (2014).

Though I will say I found it funny how you scoffed at the notion of a productive out when put the ball into play but then basically defended the notion of a productive strikeout.

Link to comment
Share on other sites

As I said in the other thread, the problem with this mentality is that when a front office goes all in with it it will lead to something like what the Braves have had for the past two seasons with a feast or famine lineup and players going into sudden decline. The high K rate is overlooked by most when they're hitting for power (2013) because they put up flashy numbers and win at a higher rate but when they do not offensive production collapses and the losses pile on (2014).

Though I will say I found it funny how you scoffed at the notion of a productive out when put the ball into play but then basically defended the notion of a productive strikeout.

I agree, a stable line-up has a cross-section of skills that work synergesticly. However, there's no evidence that a high-strikeout team is more prone to offensive decline than any other. Every hitter waxes and wanes over the course of a season and seasons. And I challenge your assumption that the Braves high-K rate was "overlooked" in 2013; there were many, many people questioning if the offense was sustainable.

Please also point to where I "defended the notion of a productive strikeout". I don't think I used the phrase or ever advanced that notion. I did say there was sometimes a correlation of better selective hitting, which brings about harder hit balls and more walks, and higher strikeouts.

Link to comment
Share on other sites

Baseball Prospectus is probably the best, but it's subscription.

Fangraphs.com has a glossary of what the stats mean, and has some good analysis as well. They also host Steamer and ZIPS projections for the next season, if you want to see what the computers think players will do.

Baseball-Reference.com is IMO the best repository of pure stats, both traditional and sabermetric.

BeyondTheBoxscore.com is a saber-centric article site hosted by SBNation. It's hit-or-miss, but sometimes has good stuff.

Quite frankly, I don't really keep up with how to calculate all this stuff since I'm not really a math guy. I just think a lot of these stats quantify performance better than most traditional stats.

ISO (Isolated Power) is essentially batting average, taking out all the singles and weighing doubles, triples, and homers. Someone with ISO in the .200 range is likely an elite power hitter.

wOBA (weighted On-Base Average) is a better version of OPS. It weighs each time a batter on base by the Linear Weight value. Its scales the same as OBP, so over .340 is good, over .380 is elite.

Edited by K26dp
Link to comment
Share on other sites

I agree, a stable line-up has a cross-section of skills that work synergesticly. However, there's no evidence that a high-strikeout team is more prone to offensive decline than any other. Every hitter waxes and wanes over the course of a season and seasons. And I challenge your assumption that the Braves high-K rate was "overlooked" in 2013; there were many, many people questioning if the offense was sustainable.

Please also point to where I "defended the notion of a productive strikeout". I don't think I used the phrase or ever advanced that notion. I did say there was sometimes a correlation of better selective hitting, which brings about harder hit balls and more walks, and higher strikeouts.

Some hitters strike out because they are taking a lot of pitches, looking for a ball to drive, and making the pitcher work for the out.

In the end, the discrete results may be the same, but which hitter is more likely to have a positive outcome his next plate appearance?

When someone says “strikeouts don’t matter”, they are really talking about the second type of batter. These kinds of batters may have high strikeout rates, but provide value because they get on base a lot (that is, they are good at not getting out in general), or hit for power, or run the bases well, or provide great defense. These types of players are All-Stars and team MVPs, even with high strikeout rates.

Basically, if you are making pitchers work more then you consider it a productive strikeout. Not that I disagree with it, in fact I completely agree, I just found it funny earlier in the post you played down the notion of a productive out when batted into play.

Link to comment
Share on other sites

Basically, if you are making pitchers work more then you consider it a productive strikeout. Not that I disagree with it, in fact I completely agree, I just found it funny earlier in the post you played down the notion of a productive out when batted into play.

I think you're missing the basic simple point of his thread. A strikeout is better than a weakly hit ball that turns into a double play. In that case "put the ball in play and good things happen." looks totally foolish.

Link to comment
Share on other sites

Basically, if you are making pitchers work more then you consider it a productive strikeout. Not that I disagree with it, in fact I completely agree, I just found it funny earlier in the post you played down the notion of a productive out when batted into play.

Ah, I see what you're saying. I wouldn't call that a "productive" strikeout, because the end result was still a failure. It more goes back to the point about hitting professional pitching being hard, and that what you want to see from a hitter in an at-bat is someone with a plan, swinging at pitches in the zone, laying off pitches out of the zone, spoiling pitcher's pitches with two strikes, and punishing mistakes. Even if a hitter does all that, he may strike out, or ground out, or whatever. Those strikeouts don't bother me that much, because you know that kind of hitter is going to have quality plate appearances.

Link to comment
Share on other sites

I think you're missing the basic simple point of his thread. A strikeout is better than a weakly hit ball that turns into a double play. In that case "put the ball in play and good things happen." looks totally foolish.

No, I get it. Not all outs when put into play are equal and not all strikeouts are equal. A GIDP is as counterproductive as a strikeout with a RISP while on the flip side a sacrifice out to advance a runner (sabermetric arguments aside) to increase the chance he'll get batted in is as productive as a strikeout where the pitcher has to work hard to get it... sure, they both end in an out and no one wants that but they bring an overall positive. That was my point.

Link to comment
Share on other sites

I agree completely that in the example of a DP a K can be more beneficial than a weak hit. It like almost everything is subjective to situation. DP I'd rather a K, no one on and no outs I hate a K. Very well written and entertaining thread.

You know what I hate with no one on and no outs more than anything? Hacking at the first pitch out of the zone and grounding out weakly to the infield.

Link to comment
Share on other sites

You know what I hate with no one on and no outs more than anything? Hacking at the first pitch out of the zone and grounding out weakly to the infield.

I hated when Braves hitters had a hitters count and swung at pitches off the plate.

I also hated when they swung at pitches low and outside and ground right to SS when all was needed was a ball to the OF for sac fly.

I hated when they tried to pull outside pitches, hitting into DPs.

MLB pitchers adapted to Braves hitters and the Braves never countered. Pitchers knew they did not have to throw strikes to get Braves batters out.

Supporting K2 original post. Strike outs are not all encompassing, they are a symptom of poor hitting. The real problems for the a braves as hitters were they lacked discipline at the plate nor did they have situational awareness. They had poor approach to too many at bats last year. Opposing pitchers knew it and took advantage of it. The braves did not adjust. Lets hope the new hitting coach can help.

Edited by AEFalcon
Link to comment
Share on other sites

Guest Homer the Brave

I hated when Braves hitters had a hitters count and swung at pitches off the plate.

I also hated when they swung at pitches low and outside and ground right to SS when all was needed was a ball to the OF for sac fly.

I hated when they tried to pull outside pitches, hitting into DPs..

You talking about Freddie Freeman here? He is a perfect example of what is wrong. When he was a fearless .300 hitter, he was spraying the ball around everywhere. Last year, he began trying to pull everything, even to the point they were putting the shift on him. What idiot hitting coach couldn't see this and correct it?

Link to comment
Share on other sites

You talking about Freddie Freeman here? He is a perfect example of what is wrong. When he was a fearless .300 hitter, he was spraying the ball around everywhere. Last year, he began trying to pull everything, even to the point they were putting the shift on him. What idiot hitting coach couldn't see this and correct it?

What idiot thinks it's as simple as saying "don't pull outside pitches?"

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...